ECOLOGY AND BIOLOGY OF THE PACIFIC WALRUS 



259 



Adjustment of the sample pregnancy rates in Table 47 is required because the 

 specimens were taken in different months, and the criteria for identification of 

 pregnancy were not uniform. "Pregnant" animals in samples B, C, and E were 

 identified by the presence of a large corpus luteum in the ovaries, because most of 

 them were taken before implantation of the blastocyst. Conversely, samples A 

 and D were taken after implantation, and the animals were identified as 

 pregnant by the presence of a conceptus. Because there is some failure in 

 fertilization, conception, and gestation, as noted earlier, the presence of a corpus 

 luteum early in pregnancy is not equivalent to presence of a conceptus later; 

 neither of these is equivalent to live birth. 



The mean difference between ovulation and conception rates has been esti- 

 mated as about 12.5 % (Table 35). I have assumed that the mortality of shed ova 

 and blastocysts is evenly distributed over a 5-month pre-implantation period 

 (1 February to 1 July) and have adjusted the pregnancy rates of samples B, C, 

 and E accordingly, to obtain the equivalent of conception rates (Table 49) . 

 Because samples A and D were taken at or soon after implantation, their 

 pregnancy rates are tantamount to conception rates and require no adjustment. 



The derived conception rates range from about 34 to 43% per year (mean, 

 38.7%). Because the mean difference between conception and birth rates is 

 about 5.2% (Table 35), the most probable adjusted birth rates of the samples 

 would range from about 32 to 41 % (mean, 36.7%) per year (Table 49). If the 

 population as a whole in 1960 included 38 % females of breeding age (Table 43) 

 its crude birth rate would have been between 12 and 16 % (mean, 14 %) per year. 

 If the population in 1972 included 46% females of breeding age, its crude birth 

 rate would have been between 15 and 19% (mean, 17%) per year. 



Table 49. Adjusted sample rates of conception and birth in samples 

 A to E of fertile females from Table 47. 



, Conception Birth 

 Sample 



Sample size No. % No. % 



A 110 41 37.2 38.9 35.4 



B 49 16.6 33.9 15.7 32.0 



C 171 60.4 35.3 57.2 33.4 



D 201 87 43.3 82.5 41.0 



E 34 13.9 40.9 13.2 38.8 



Total 565 218.9 38.7 207.5 36.7 



Survival and Recruitment 



The unusually low reproductive rate of walruses must be balanced by lower 

 mortality and a higher recruitment to breeding age than in other pinnipeds. The 

 unusually long period of parental care (2 years) probably is a major factor 

 contributing to that high survivorship. Loughrey (1959) estimated that re- 

 cruitment to breeding age in the Canadian walrus population amounted to about 

 10% per year; Mansfield (1966) implied that recruitment was somewhat greater 



